---
license: mit
language:
- de
pipeline_tag: text-classification
metrics:
- f1
library_name: transformers
---

# PopBERT

PopBERT is a model for German-language populism detection in political speeches within the German Bundestag, based on the deepset/gbert-large model: https://huggingface.co/deepset/gbert-large

It is a multilabel model trained on a manually curated dataset of sentences from the 18th and 19th legislative periods.
In addition to capturing the foundational dimensions of populism, namely "anti-elitism" and "people-centrism," the model was also fine-tuned to identify the underlying ideological orientation as either "left-wing" or "right-wing."

# Prediction

The model outputs a Tensor of length 4.
The table connects the position of the predicted probability to its dimension.

| **Index** | **Dimension**            |
|-----------|--------------------------|
| 0         | Anti-Elitism             |
| 1         | People-Centrism          |
| 2         | Left-Wing Host-Ideology  |
| 3         | Right-Wing Host-Ideology |

# Usage Example

```python
from transformers import AutoModel
from transformers import AutoTokenizer

# load tokenizer
tokenizer = AutoTokenizer.from_pretrained("luerhard/PopBERT")

# load model
model = AutoModel.from_pretrained("luerhard/PopBERT")

# define text to be predicted
text = (
    "Das ist Klassenkampf von oben, das ist Klassenkampf im Interesse von "
    "Vermögenden und Besitzenden gegen die Mehrheit der Steuerzahlerinnen und "
    "Steuerzahler auf dieser Erde."
)

# encode text with tokenizer
encodings = tokenizer(text)

# predict
with torch.inference_mode():
    out = model(**encodings)

# get probabilties
probs = torch.nn.functional.sigmoid(out.logits)
print(probs.detach().numpy())
```

```
array([[0.87651485, 0.34838045, 0.983123  , 0.02148381]], dtype=float32)
```


# Performance

To maximize performance, it is recommended to use the following thresholds per dimension:

```
[0.415961, 0.295400, 0.429109, 0.302714]
```

Using these thresholds, the model achieves the follwing performance on the test set:

| Dimension           | Precision     | Recall        | F1            |
|---------------------|---------------|---------------|---------------|
| Anti-Elitism        | 0.81          | 0.88          | 0.84          |
| People-Centrism     | 0.70          | 0.73          | 0.71          |
| Left-Wing Ideology  | 0.69          | 0.77          | 0.73          |
| Right-Wing Ideology | 0.68          | 0.66          | 0.67          |
| ---                 | ---           | ---           | ---           |
| micro avg           | 0.75          | 0.80          | 0.77          |
| macro avg           | 0.72          | 0.76          | 0.74          |
